0
$\begingroup$

When we calculate lower and higher frequency such that power becomes half in LCR circuit we end up with two value. For a complete derievation please see here. I have copied relevant information below:

  1. Resonant frequency:

    \begin{align} X_L &= X_C \\ \omega_r L - \frac{1}{\omega_r C} &= 0 \\ \omega_r &= \frac{1}{\sqrt{LC}} \end{align}

  2. Current: At frequency $\omega_r$, $Z_T$ is minimized and $I_S$ is maximized. \begin{align} I_\text{max} = \frac{V_\text{max}}{Z} = \frac{V_\text{max}}{\sqrt{R^2 + (X_L - X_C)^2}} = \frac{V_\text{max}}{\sqrt{R^2 + \left(\omega_r L - \frac{1}{\omega_r C} \right)^2}} \end{align}

  3. Lower cut-off frequency $\omega_L$: At half power $P_m/2$, $I = I_m / \sqrt 2$ and $Z = \sqrt{2}R$. Then $X=-R$ (capacitive) and so $$\omega_L = -\frac{R}{2L} + \sqrt{\left(\frac{R}{2L}\right)^2 + \frac{1}{LC}}$$

  4. Upper cut-off frequency $\omega_U$: Again $I = I_m/\sqrt{2}$ and $Z=\sqrt{2}R$, but this time $X=+R$ so $$\omega_U = +\frac{R}{2L} + \sqrt{\left(\frac{R}{2L}\right)^2 + \frac{1}{LC}}$$

  5. Bandwidth: $$(\omega_L + \omega_U)/2 = \sqrt{\left( \frac{R}{2L}\right)^2 + \frac{1}{LC}}$$

Question: Does this mean bandwidth is not centered at the resonant frequency in an LCR series circuit?

$\endgroup$
4
  • $\begingroup$ There is no reason why a resonance curve should be symmetrical about the resonant frequency. $\endgroup$
    – Farcher
    Commented Feb 15 at 9:19
  • $\begingroup$ But we often write upper frequency = resonant frequency + R/2L and lower frequency = resonant frequency - R/2L $\endgroup$
    – ssr
    Commented Feb 15 at 9:29
  • $\begingroup$ Also every graph shows them symmetric about resonant frequency, I have added one of them at last in question $\endgroup$
    – ssr
    Commented Feb 15 at 9:38
  • $\begingroup$ I've edited this question to use mathjax instead of images. Please always use mathjax for math on this site. $\endgroup$
    – DanielSank
    Commented Feb 15 at 20:13

1 Answer 1

1
$\begingroup$

Does this mean bandwidth is not centred at resonant frequency in LCR series circuit? If not, please explain.

It IS centered but on a logarithmic frequency axis, not a linear one. The center frequency is the geometric mean of the half-power frequencies, not the arithmetic mean

But we often write upper frequency = resonant frequency + R/2L and lower frequency = resonant frequency - R/2L

That's incorrect or, at best, an approximation for a relatively small bandwidth. We have

$$\frac{\omega_0}{\omega_{low}} = \frac{\omega_{high}}{\omega_{0}}$$ but $$\omega_{high} - \omega_{0} \neq \omega_0 - \omega_{low}$$

The half power frequencies are determined as the frequency where the reactance has the same magnitude as the resistance. Below the resonance, the impedance is dominated by the capacitor, i.e. the reactance is negative. Above the resonance its inductive, i.e. the reactance is positive.

We get a solution for each case

$$ \omega_{high} = \sqrt{\omega_0^2 + \frac{R^2}{4L^2}} + \frac{R}{2L} $$ $$ \omega_{low} = \sqrt{\omega_0^2 + \frac{R^2}{4L^2}} - \frac{R}{2L} $$

The bandwidth is the difference, which is $$ B = \omega_{high} - \omega_{low} = \frac{R}{L} $$

The product is $$ \omega_{high} \cdot \omega_{low} = \omega_0^2 + \frac{R^2}{4L^2} - \left( \frac{R}{2L} \right)^2 = \omega_0^2 $$

That means that resonance frequency is the geometric mean of the half-power frequencies.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.